개인화된 도움을 위한 증강현실기반 메시지 주석시스템
Nguyen Van Vinh
†․전 희 성
††요 약
본 연구에서는 특정개인에 맞춘 정황정보를 제공할 수 있는 개인화 메세지 주석시스템을 제안한다. 본 시스템에서 컨텍스트는 사용자의 신 분, 위치, 시간등을 포함하는 실체로 정의하며 사용자의 정체는 개인화된 메세지 생성의 주된 데이터이다. 컨텍스트를 감지하기 위해 증강현실 기술이 사용되며, 사용자의 위치를 추적하기 위해 실제 물체에 마커가 부착된다. 증강현실 기술은 인간의 지각과 상호작용을 강화시키는 아주 효과적인 주석 방법으로 사용될 수있다. 메시지는 가상 문자메세지 또는 3차원 가상물체일 수 있다. 실험결과는 제안된 시스템이 실시간으로 효과적으로 작동됨을 보여주며 또한 개인화 메시지를 제공하는 모바일 서비스로 사용될 수 있음을 보였다.
키워드 : 증강현실, 상황인식 컴퓨팅, 모바일 서비스, 개인화 메세지
AR-based Message Annotation System for Personalized Assistance
Nguyen Van Vinh
†․Jun, Heesung
††ABSTRACT
We propose an annotation system, which allows users moving on an environment to receive personalized messages that are generated by exploiting contextual information. In the system, the context is defined as an entity including user’s identity, location and time. Identity of user is a key data to enable personal aspect of generated message. For sensing the context, the proposed system uses AR(augmented reality) technology. Markers are attached to real objects for tracking user’s location. AR can provide an effective annotating method to enhance human’s perception and interaction abilities. The received message can be a virtual post-it or three-dimensional virtual model of object overlaid onto the real-world view. Experimental results show that the proposed system works well in real-time with high performance and it can be used as a mobile service for personalized messaging.
Keywords : Augmented Reality, Context-aware Computing, Mobile Service, Personalized Messaging
1. Introduction
1)Context-aware computing has been a hot research topic and it greatly contributes to the enhancement of our life’s quality. Many context-aware applications and systems have been developed to make our life more convenient and comfortable[1]. Examples are travel infor- mation while travelling[12], local shopping guide[11], the nearest parking lot when needed to stop your car, up- to-date local news, etc. Those applications are developed by taking advantage of contextual information such as location, time, user’s activity, etc. However, most of them
※ 이 논문은 2004년 울산대학교의 연구비에 의하여 연구되었음.
†준 회 원 : 세종대학교 디지털콘텐츠학과 박사과정
††정 회 원 : 울산대학교 컴퓨터정보통신공학부 교수(교신저자) 논문접수 : 2009년 9월 16일
수 정 일 : 1차 2009년 11월 2일 심사완료 : 2009년 11월 2일
provide helpful information in public way and also they
do not exploit user’s profile that is considered as an
important data for the present business. With the rapid
development of mobile devices such as wearable computer,
personal digital assistant (PDA) or cell phone, data services
tend to be personalized. Research and development for
mobile services focus on the personal aspect of mobile
devices. CHIP (Cultural Heritage Information Personalization)
project [4] provides personalized access to combined
cultural heritage content. In CHIP, the personalization that
is based on a user model containing the interests, goals,
background and knowledge of the user, improves museum
website usability by supporting user navigation and
assisting them in finding appropriate and interesting
information. A context-based information sharing system,
cPost-it, allows users to share various types of media
data by exploiting context such as the user’s identity
(ID), location, time, etc[5]. In order to offer personalized services in a consistent manner, a Personal Service Environment[6] is presented that besides profile manage- ment, provides generic service discovery, content adaptation and service adaptation functionality. Jorstad et al.[7] present an analysis of the requirements for service personalization and proposes a generic service architecture that supports personalization. Mechanisms for efficient synchronization of profiles based on semantic distinctions are introduced in[8].
In this paper, we focus on the personal aspect of information retrieving on mobile devices. The proposed annotation system allows users to see not only general information in public places, but also private messages according to the user’s identity. For example, a big university has many buildings and departments so that a lot of public bulletin boards are available. Most information in these boards is too general or may be not specific to users. Also the information may be degraded due to environmental effects. In order to give more information (private messages, comments, virtual models, etc) to a specific user, an annotation system is proposed. A user can take his PDA or mobile computer equipped a camera to go anywhere around the university. Annotation system will track user’s location using markers which have been attached to bulletin boards for place definition. For each board, not only information written on the board can be perceived, the user also can see private message in terms of text or thee-dimensional (3-D) virtual objects overlaid on board’s marker. For unknown(or outside) user of a department, a general message can be created and displayed such as “Welcome to the exhibition of our department at room 317” or “A new model of car has just been created. Please see this demo” with a 3-D virtual model of car shown over the marker. Users of the department can get private messages like “We have meeting at room 7-314” or “Please meet Mr. Kim at 2pm” that is made according to user’s position (manager of department, professor, student).
In such annotation systems, Augmented Reality (AR) is an effective means to enhance human’s perception and interaction abilities. AR refers to a field of computer research which deals with the combination of real world and computer generated data. Most AR research is concerned with the use of live video imagery which is digitally processed and “augmented” by the addition of computer generated graphics. For annotation applications, AR is normally used to provide users with helpful information overlaid onto the real-world view.
The paper is structured as follows. In section II, we introduce context-aware mobile computing, AR technology, and their related works. Section III presents detailed design of the proposed system. Experimental results will be shown in section IV. Finally, we conclude this paper and discuss future works in section V.
2. Related Works
2.1 Context-aware mobile computing
Context-aware mobile computing aims to discover and take advantage of contextual information. Many researchers have studied this topic and built several context-aware applications to demonstrate the usefulness of this new technology. The Active Badge[9] system is generally considered to be one of the first context-aware appli- cations. With the system persons could be located in an office and calls forwarded to the closest phone. The office personnel wore badges that transmitted IR-signals (so-called active badges). A network of sensors placed around the office building picked up the signals and a central location server polled these sensors. The telephone recep- tionist could find out where a person was and direct the call to an appropriate phone. Based on Active Badge, Teleporting[10] is built as a tool to dynamically map the user interface onto the resources of the surrounding computer and communication facilities. The Shopping Assistant[11] can guide the shoppers through the store, provide details of items, help locate items, point out items on sale, and so forth. Cyberguide[12] is a tourist guide system which provides information services to a tourist about his current location. The Conference Assistant helps conference attendees by using a variety of context infor- mation[13]. It examines the conference schedule, topics of presentations, user’s location and user’s research interests to suggest the presentations to attend. The Office Assistant[14] is an agent that interacts with visitors at the office door and manages the office owner’s schedule.
The ComMotion project[15] provides a service that each reminder message is created with a location and it is delivered via voice synthesis when the intended recipient arrives at that location.
As mentioned at the previous section, those applications
are not personalized. Helpful information is mainly based
on user’s location which can be sensed by using several
techniques. One location-tracking technique is the Global
Positioning System (GPS). However, GPS cannot work in
an indoor environment due to building’s obstructive affect
on GPS signal strength. Thus, other techniques based on
infrared (IR)[12] or radio frequency (RF)[11] have been proposed. To reduce the cost of system and to adapt with AR environment, the proposed system is based on image processing for tracking user’s location.
2.2 Augmented Reality
AR refers to the real-time insertion of computer- generated graphical content into a real scene. The AR environment is a real-world based environment where physical entities and computer-generated entities coexist.
R. T. Azuma[1] has classified applications for which AR is applied. In medical applications, an AR system can support general medical visualization tasks in the surgical theater[16-17]. For annotation and visualization, information helpful to the user can be overlaid onto the real-world view[18-19]. In manufacturing and repair, AR can be used to support the assembly, maintenance, and repair tasks of complex machinery[20-21]. Moreover, AR can aids user in testing the visual robot’s action and using the real robots. Entertainment and military aircraft also are suited areas of AR applications which can place user into environments with intelligent virtual creatures[22] or provide soldiers with simple task-specific systems by the
“smart clothing” method[23]. AR applications including mobile, collaborative, and commercial are discussed in a survey paper[2]. The Touring Machine is a tour guide system using wearable devices which presents infor- mation about Columbia University’s campus[24]. The Déjà vu[25] is a navigation system in the indoor environment using a mobile computer to take current position with GPS sensor and a CCD camera. Billinghurst’s Shared Space work showed how AR can be used to seamlessly enhance face to face collaboration[26]. Henrysson et al.[27]
demonstrate the first face to face collaborative game running on a mobile phone. For commercial applications, AR has been used for real-time augmentation of broadcast video to enhance sporting events and to insert or replace advertisements in a scene[28]. The Fig. 1 shows some examples of AR applications. In the first picture, a virtual pre-display of the pipes on the screen of an AR system could help a plumber to install the pipe in the most efficient way. Next, a virtual post-it is displayed to provide detailed information of a person while a user is moving in unknown environment. For car racing, AR systems are used for providing fans “not at the track”
with graphic effects and relevant data enabling them to follow the action of their favorite drivers more easily. In soccer games, virtual offside line created by an AR system enables the viewer see exactly when and how the foul was committed.
(a)
(b)
(c)
(Fig. 1) Examples of AR applications. (a) Industrial AR application (Siemens Corporate Research[29]). (b) Annotation application (Personal Identity System[19]). (c) Consumer AR application[28]
3. Proposed System
3.1 System Overview
The proposed system is designed in client-server architecture. It consists of a portable computer (a PDA or a mobile PC with a wearable camera) connected to server via wireless network. The system uses camera to capture image sequences of real world. Portable computer process those images as input to detect, track and recognize markers attached to real objects. Afterwards, the data including user ID and marker ID will be sent to application server. At server, marker ID is mapped to user’s location. Then, messages will be retrieved from database based on user context: ID, location, time.
Messages can be texts or 3-D graphic objects defined in VRML. Portable computer displays text message to the user in the form of virtual post-it attached to the top-left corner of live video screen whereas 3-D virtual symbol can be overlaid on the marker.
As mentioned above, the system uses marker to track user’s location (e.g. which department or bulletin board).
In such real-time distributed application, a challenge is to
maintain high performance due to the expensive compu-
tation of image processing and the delay of client-server
data transmission. Hence, simple markers are employed to
reduce the cost of computation in detection, tracking and
Video stream from camera
Marker Detection Message Rendering
Marker Tracking
Location Mapping Context Manager
Database Client
Server
Marker Recognition
Login
Authentication User input
(Fig. 3) Process diagram of the system
Image sequences
Binarization Marker
Dectection
Square verification Labeling Contour Detection
Marker Recognition
Line Contour Estimation Corner Detection Pattern Normalization
Template Matching
(Fig. 4) The Flow diagram of marker detection and recognition processes
recognition processes. Markers consist of the black and
white squares as shown in Fig. 2. The predefined patterns make the difference of markers.
(Fig. 2) Four examples of marker used in the system
3.2 Annotation system
A process diagram of the system is shown in Fig. 3.
Given an image sequence as input, the system generates personalized messages and shows them to the user. The proposed system comprises eight processes: the marker detection, tracking, recognition, message rendering, login, on the client and authentication, location mapping, context manager on the server.
Authentication is to obtain personal property of message. A user having his own profile data such as name, job, position can get a specific message upon his current situation like standing location or current time.
Each person can be uniquely determined by a user’s identity that is easily attained via registration process provided by service provider. In the beginning, a user enters his ID and password. The data will be sent to server and be processed for checking validity. Then, the user’s identity is given to other processes as input and this information is frequently used during data trans- mission between client and server for personalizing data.
Computer vision algorithms are applied in marker detection and marker recognition process[30]. Marker detection process is to detect potential markers in image sequences taken from camera. It is based on a contour
detection technique to find objects having shapes similar to a marker whereas marker recognition is based on corner detection and template matching to identify a marker. The detailed steps are shown in Fig. 4.
The system performs the image binarization and then detects the contour of objects in the image. Gray-level images are converted to black and white images by binarization. The binarized image is processed by labeling algorithm which helps to distinguish separated objects.
After labeling, contour following is performed to extract contours of objects. For each contour, the system uses square verification method to check the similarity of their shapes and a rectangle.
Marker recognition uses template matching technique to recognize markers. First, line contour estimation is applied and then the process detects sub-pixel corners of the found contour. Afterward, the region of marker-considered object is extracted using detected corners. The extracted image is then normalized to the predefined size of pattern. Finally, the system uses a correlation matching method to evaluate the marker similarity. The detected marker is compared with all predefined patterns, and a matching score computed through the correlation is returned. The detected marker is then recognized as the one giving the highest cumulative score.
Whenever a marker is recognized, its information will
be sent to marker tracking. This process is to keep track
of detected marker with its information including ID
number, position, size and direction. When the camera is
rotated or moved, the marker information is frequently
Internet
Web Server
Database Administrator
User
User - manage messages (insert/update/delete) - maintain marker and its relation to location
- register account - update profile ...
(Fig 5) Diagram of remote data manage
updated. In order to make functionality of annotation
properly, marker tracking also have to manage perso- nalized messages which are stored on a remote database.
Each time a marker is identified, the process determines the corresponding message of the marker and then sends those data to message rendering for annotation. For getting messages at remote database, marker tracking need to make a request to application server by sending information including marker ID and user’s identity. A problem occurring in this process is that consecutive data transmission between client and server affects the performance of the system. To solve that problem, the payload of client-server communication need to be decreased and the client sends requests to server only if needed. To do this, message tracking is implemented. Not only current marker is monitored but also previous detected markers and their corresponding messages are tracked. Once a relevant message of a marker has been retrieved, it can be reused for that marker during the user’s session. The system checks the current detected marker’s appearance and if this is not the first time, previous received message will be used.
Getting data from client as input, location mapping connects to remote database for determining location. The relationship between marker and location is predefined and is stored in database in term of relation of tables.
One marker is mapped to one location uniquely. All data is passed to context manager for next processing.
The purpose of context manager is to extract perso- nalized messages upon received data. The system uses 3W (Who, Where, When) as a context to provide the proper service[31]. “Who” is a user’s identity which represents the name of the user or group the user belongs. The system classifies messages based on identity. “Where” represents the location of user and
“When” is information about the current time of message request. Given “Who”, “Where” and “When” as input, Context Manager connects to the database for messages retrieving. More detail of data organization and management is explained in next section.
3.3 Data Management
As mentioned above, data is mainly stored in a remote database. The data includes information about user, marker, location, and message. User-related data consists of user profile (name, job, work place, position, etc) and account (ID, password, role). User account is necessary for authentication and data management.
A marker has its own ID, name and pattern file used
for recognition. To improve the performance, marker recognition needs to be designed as a client process.
Pattern files of marker are stored not only in the database, but also in the client. Thus, a synchronization mechanism is demanded for matching pattern files between client and database. A client automatically updates new patterns from database after successful login. Because pattern file is ASCII file containing pixel values of normalized image of marker template, the size of file is normally small. Moreover, the number and model of marker used in system is generally stable. Hence, the task of synchronization does not take much time for updating and also does not affect the system performance.
In the system, location means the place where user is standing. It is “Where” playing important role in message extraction. User location is determined and tracked by marker. The system assigns a marker to one location so that each detected marker in client will be mapped to a corresponding location. This relationship can be defined in linked table on the database.
Stored message is personalized information created to inform user an event or provide more detail about objects.
The content of message varies frequently according to user situation such as who he is, where he is standing and what time is. Message can be a string that provides information to user or a 3-D virtual object defined by VRML. It is easy to model a real object or an animation of world by VRML. Therefore it can enhance user’s perception and interaction ability. Message will be displayed to user in term of virtual post-it in case of its string format. Otherwise, a virtual model of object will be overlaid on marker.
To manage data, a web-based remote control service is provided. User of this service can be classified into two groups: administrator and normal user. Fig. 5 depicts a general view of remote data management.
A normal user needs the service to register an account
Virtual post-it message 3D virtual object overlaid on marker
Previous messages
(Fig. 6) Implemented PMA client - main view form with example of augmented messages
(Fig. 7) The result of personalized messaging assistance:
(top-left-no message, top-right-virtual post-it notice, bottom-left-3D object, bottom-right-3D object and explanation)
and update information of his profile as well. An administrator has a more complicated task. He initiates messages then assigns them to each user in relation with location and time. A message can be assigned a start time and an expired time for limited access. If the time of request is within this period, the message will be provided to user. Otherwise, user cannot receive the message. Another important mission of administrator is to manage information of location, marker and maintain their relation. Location data is created for each real place(e.g.
building, department, floor, etc). Also, marker data is made and handled in relation with each location.
4. Experimental Results
A test version of the proposed system has been built using Visual C++ 6.0 mostly based on ARToolkit open source library[30]. The interface of Personalized Messaging Assistant (PMA) client is shown in Fig. 6.
With login form, users input their information including user name and password for authentication. The data is known in advance for each person via registration process. The main view form contains a window of live video that shows image sequences continuously captured from real world. In addition, a historical message board is available that enables user to see previous received messages. A virtual post-it is displayed on the top-left corner of live video window for a text message whereas a 3D virtual model of car is overlaid on a marker.
In order to evaluate the performance, the system has been tested in our university. The client program has been installed in notebook with a USB digital camera and a wireless LAN card. For location tracking, markers have
been attached in several places such as bulletin board, room door, wall, etc. A user taking the mobile PC with camera walks around the floor. The image sequences are captured continuously by the camera. Once a marker is detected and then recognized, the predefined relevant message is displayed to the user. Fig. 7 shows examples of personalized messaging assistance. Four persons using PMA system stand in front of a bulletin board and direct their camera to the same marker attached on the board.
The captured screens on their PMA client are shown in the figure respectively. The first person got no message.
The second person received a virtual post-it notice. The third person saw a 3D virtual arm of robot whereas the last person observed 3D virtual car and an explanation.
The problem encountered in the test is the occurrence of delay of video stream. In the two first cases, the messages were displayed to the users immediately.
However, last two persons have observed the delay of video stream when the marker is recognized at the first time. This problem is caused by the transmission time of files between server and client. In case of text message, the small amount of data needed to transfer makes the time of data retrieving very short. Since text message can be transferred very quickly, there is no delay. But the transferring VRML files needs more time due to the noticeable size of file. Consequently, retrieving process makes the delay problem with respect of 3D virtual object message. Table I lists the tested VRML files with their sizes and corresponding delay times.
However, the proposed system uses message tracking
mechanism so that the delay only occurs at the first
recognition time of each marker. Hence, the performance
File name Size (KB) Time delay (ms)
Ufo.wrl 359 1243
Car.wrl 264 863
GrayPipe.wrl 163 521
Snowman.wrl 5.66 34
<TABLE 1>
VRML Files and Time Delay
of the system is not affected by the delay problem.
5. Conclusion
In this paper, an annotation system for personalized messaging is proposed. The key contribution of this work is to design a system which is a combination of context- aware based mobile computing, private messages and augmented reality. The experimental results demonstrate the possibility of real-time processing. Thus, the system suggests a realizable mobile service for personalized messaging. In the future, we will implement the system on PDA for further evaluation of performance.
References
[1] G. Chen, D. Kotz, “A Survey of Context-Aware Mobile Computing Research,” Technical Report, TR2000-381, Dept.
of Computer Science, Dartmouth College, November, 2000.
[2] R. T. Azuma, “A survey of Augmented Reality,” Teleope- rators and Virtual Environments, Vol.6, No.4, pp.355-385, Aug., 1997.
[3] R. Azuma, et al.., “Recent advances in augmented reality,”
IEEE Computer Graphics and Applications, Vol.21, No.6, pp.
34-47, Nov/Dec., 2001.
[4] CHIP project website: http://www.chip-project.org [5] S.Jang, S.Lee, W.Woo, “cPost-it: Context-based Information
Sharing System,” Lecture Note Computer Science, Vol.2899, pp.352-363, 2003.
[6] M.M. Lankhorst, H. van Kranenburg, A. Salden,, A.J.H Peddemors, “Enabling Technology for Personalizing Mobile Services,” Proceedings of the 35th Annual Hawaii Inter- national Conference on System Sciences (HICSS'02), Vol.3, Jan., 2002.
[7] I. Jorstad, D. van Thanh, S. Dustdar, “The Personalization of Mobile Services,” Proceeding of the IEEE International Conference on Wireless and Mobile Computing, pp.473-479, Aug., 2005.
[8] O. Drogehorn, B. Wust, K. David, “Personalised applications and services for a mobile user,” Proceedings Autonomous Decentralized Systems, ISADS 2005, pp.473-479, April 2005.
[9] R. Want, A. Hopper, V. Falcao J. Gibbons, “The Active Badge
Location System,” ACM Transactions on Information Systems 10(1) pp.91-102.
[10] F. Bennett, T. Richardson, and A. Harter., “Teleporting- making applications mobile,” In Proc. of the IEEE Workshop on Mobile Computing Systems and Applications, pages 82-84, Santa Cruz, California, Dec., 1994. IEEE Computer Society Press.
[11] A. Asthana, M. Cravatts and P. Krzyzanowski, "An Indoor Wireless System for Personalized Shopping Assistance,"
Proceedings IEEE Workshop on Mobile Computing Systems and Applications, Santa Cruz, CA, Dec. 8-9, 1994.
[12] G. D. Abowd, C. G. Atkeson, J. Hong, S. Long, R. Kooper, and M. Pinkerton, “Cyberguide: A mobile context-aware tour guide,” ACM Wireless Networks, pages 421-433, 1997.
[13] A. K. Dey, D. Salber, G. D. Abowd, and M. Futakawa, “The conference assistant: Combining context-awareness with wearable computing,” In Proceedings of the 3rd International Symposium on Wearable Computers (ISWC '99), pages 21-28, San Francisco, CA, October 1999.
[14] H. Yan and T. Selker, “Context-Aware Office Assistant,”
In Proceedings of the 2000 International Conference on Intelligent User Interfaces, pages 276-279, New Orleans, LA, January 2000. ACM Press.
[15] N. Marmasse and C. Schmandt, “Location-aware Information Delivery with comMotion,” Proc. 2nd Int’l Symp. Handheld and Ubiquitous Computing (HUC 2K), Lecture Notes in Computer Science, P. Thomas and H.-W. Gellersen, eds., Vol.
1927, No.1, Springer-Verlag, Berlin, 2000, pp.157-171.
[16] M. Bajura, H. Fuchs, and Ohbuchi, “Merging virtual objects with the real world: seeing ultrasound imagery within the patient,” Computer Graphics, pp.203-210, July, 1992.
[17] W. Birkfellner, et al., “A head-mounted operating binocular for augmented reality visualization in medicine-design and initial evaluation,” IEEE Trans. on Medical Imaging, Vol.21, No.8, pp.991-997, Aug., 2002.
[18] S. Mann, “Smart Clothing: wearable multimedia computing and personal image to restore the technological balance between people and their environments,” ACM multimedia, pp.163-174, Nov., 1996.
[19] J.B. Kim, “Personal Identity Annotation Overlay System using a Wearable Computer for Augmented Reality,” IEEE Transaction on Consumer Electronics, Vol.49, No.4, pp.1457- 1467, 2003.
[20] M. Tucervan, et al., “Calibration requirements and procedures for a monitor-based augmented reality system,” IEEE Trans. on Visualization and Computer Graphics, Vol.1, No.3, pp.255-273, Sep., 1995.
[21] Metaio website: http://www.metaio.com.
[22] A. L. P. Albuquerque, J. Gomes, and L. Velho, “An overview on virtual sets,” Lecture Notes in Computer Science, Springer-Verlag, Vol.1834, pp.91-98, 2000.
[23] http://www.ait.nrl.navy.mil/vrlab/projects/BARS/BARS.
html
[24] S. Feiner B. MacIntyer, T. Hollerer, and A. Webster, “A Touring Machine: prototyping 3D mobile augmented reality systems for exploring the urban environment,” In Proc. Int.
Sym. On Wearable Computing, pp.74-81, Oct., 1997.
[25] I. Takuya, K. Yasuyuki, “Déjà vu-An augmented reality system with mobile computer,” IPSJ SIGNotes Audeo Visual and Multimedia information processing, No.015, 2001.
[26] M. Billinghurst, I. Poupyrev, H. Kato, and R. May, “Mixing Realities in Shared Space: An Augmented Reality Interface for Collaborative Computing,” In Proceedings of the IEEE International Conference on Multimedia and Expo (ICME 2000), July 30th-August 2, New York.
[27] A. Henrysson, M. Billinghurst, M. Ollila, “Face to Face Collaborative AR on Mobile Phones,” ISMAR 2005:80-89.
[28] http://www.sportvision.com
[29] N. Navab. B. Bascle, M. Appel, and E. Cubillo, “Scene augmentation via the fusion of industrial drawings and uncalibrated images with a view to marker-less calibration,”
IEEE and ACV Int. Workshop on Augmented Reality, pp.
125-133, 1999.
[30] ARToolKit website: www.hitl.washington.edu/artoolkit [31] S. Jang, W. Woo, “ubi-UCAM: A Unified Context-Aware
Application Model,” LNAI(Context03), pp.178-189, 2003.
Nguyen Van Vinh
e-mail : [email protected]
2004년 Graduated from Hanoi University of Technology, Hanoi, Vietnam (Bachelor’s degree)
2007년 Graduated from University of Ulsan, Ulsan, Korea(Master)
2008년～현 재 : 세종대학교 디지털콘텐츠학과 박사과정 관심분야 : Computer Vision, Augmented Reality, Human
Computer Interaction
전 희 성
e-mail : [email protected] 1981년 서울대학교 전기공학과(학사) 1983년 서울대학교 대학원 전기공학과
(공학석사)
1992년 미국 Rutgers University 컴퓨터공 학과(공학박사)
1983년～1986년 금성반도체(주) 주임연구원
1992년～1993년 삼성전자(주) 통신연구소 수석연구원 2001년 미국 University of California, Irvine 방문교수 1993년～현 재 울산대학교 컴퓨터정보통신공학부 교수 관심분야 : 컴퓨터비전, 영상처리, 증강현실 등
